Escolar Documentos
Profissional Documentos
Cultura Documentos
Keywords:Non-condensable gases, Single-flash geothermal power plants, Gas removal systems, Thermodynamic analysis,
modelling and simulation of geothermal power plants.
ABSTRACT
Geothermal fluids contain Non-Condensable Gases (NCGs) at various amounts. The presence of NCGs in geothermal steam results
with a dramatic decrease in net power output increasing condenser pressure and total auxiliary power consumption. Hence, NCGs
should be withdrawn by a gas removal equipment to improve the performance of geothermal power plants (GPPs). The flashed-
steam GPPs are a relatively simple way to convert geothermal energy into electricity when the geothermal wells produce a mixture
of steam and liquid.
The primary aim of the study is to model and simulate single-flash GPPs to examine the thermodynamic performance of NCG
removal systems, which are major concerns at planning and basic design stages of GPPs. Four different NCG removal systems,
which are two-stage steam jet ejector system, two-stage hybrid system, two-stage compressor system and reboiler system are
studied. The model is validated comparing model output with Kizildere GPP output, classified as deterministic and static. In this
study, the net power output and specific steam consumption of a single-flash GPPs are evaluated depending on the separator
pressure, condenser pressure, NCG fraction and wet bulb temperature of the environment.
1. INTRODUCTION
Geothermal power can be produced by dry steam, flashed-steam, binary and Kalina plants depending on the temperature and state
of the geothermal fluid. Flashed-steam (single and double-flash) GPPs are the most commonly used power generation systems with
a total share of 61% within the installed capacity in the World, mainly because most geothermal reservoirs are formed by liquid
dominated hydrothermal systems (Bertani, 2010).
Geothermal steam, which flows through the entire cycle of conventional (dry and flashed-steam) GPPs, contains higher
concentrations of noncondensable gases (NCGs) (CO2, H2S, NH3, N2, CH4 etc.) compared with conventional fossil-fueled power
plants. The amount of NCGs contained in geothermal steam has significant impact on the power production performance of a GPP.
The NCGs in geothermal steam interfere with heat transfer in the condenser by forming a gas-blanketing effect, which raises the
condenser temperature and back-pressure on the turbine, reducing its output. In practice, the gas effect can only be overcome by
evacuating them, along with a portion of steam (Vorum and Fritzler, 2000). The power needed to extract the NCGs from the
condensers and exhaust them to the atmosphere or an abatement system is supplied from the generated electricity or by steam gas
exhausters or a combination of these; this seriously impairs the power production performance (Duthie and Nawaz, 1989). NCGs
also decrease the exergy of the fluid reducing the available work in the plant. Thus, evaluation of the net work of the turbine should
consider the NCG content (Montero, 1990). Comparing with fossil-fuelled power plants, GPPs require larger capacity NCG
removal systems which occupy a large portion in total plant cost. Therefore, selection of NCG removal system becomes a major
concern at planning and basic design stages which aim to maximize net power output and minimize both investment and operation
and maintenance (O&M) costs of GPPs in a long-term perspective (Tajima and Nomura, 1982; Hankin et al., 1984).
Changes in the resource are usually accompanied by changes in the NCG flow. Depending on the resource, the fraction of the
NCGs varies over the World from almost zero to as much as 25% by weight of steam (Hall, 1996; Coury et al., 1996). Because of
the elevated NCG levels, GPPs require large capacity NCG removal systems which play a vital role in power generation occupying
large portion in its total plant cost and total auxiliary power consumption. Therefore, selection of NCG removal system becomes a
major concern at planning and basic design stages of geothermal power plants (Tajima and Nomura, 1982; Hankin et al., 1984).
The conventional gas removal systems used in geothermal power plants are;
Jet ejectors, e.g. steam jet ejectors, which are suitable for low NCG flows (<3%) (SJEs),
Liquid ring vacuum pumps (LRVPs),
Roto-dynamic, e.g. radial blowers, centrifugal compressors, which are mainly used for large flows of NCG (>3%),
Hybrid systems (any combination of equipment above).
Besides, innovative upstream reboiler systems are another approach to remove NCGs from geothermal steam before they enter the
turbine. Recently, in GPPs hybrid NCG removal system (SJE and LRVP) are most common.
The performance of a geothermal power cycle is influenced by geothermal fluid properties such as temperature, pressure, NCG
fraction, separator/condenser pressure (Swandaru, 2006; Siregar, 2004), and wet bulb temperature of environment (Swandaru,
2006). In a specific field, temperature and pressure do not change much in the short-medium term, whereas NCG fraction may vary
significantly (Bidini et al., 1999).
1
Yildirim and Gokcen
This study examines the performance of a single-flash GPP for four different conventional gas removal options, under various
separator pressures (100-1000 kPa), NCG fractions (0-20%) and wet bulb temperatures (5-25C). The single-flash GPP is modelled
by a code written in Engineering Equation Solver (EES) (F-Chart, 2014) for two-stage steam jet ejector system (SJES), two-stage
hybrid system (HS), which consists of steam jet ejector and LRVP, two-stage compressor system (CS) and reboiler system (RS).
Geothermal fluid which is a mixture of liquid, water vapor and NCGs at the wellhead is separated into the steam and liquid phases
at the separator. Steam phase is directed to the turbine contains water vapor and NCGs. A demister is employed prior to the turbine
to remove the condensate from the steam and make sure dry steam is introduced to the turbine. After passing the turbine; steam,
condensate and NCGs ow to the condenser. The primary purpose of the condenser is to condense the exhaust steam leaving the
turbine. NCGs are accumulated and extracted by a gas removal system from the condenser. The rest is pumped to the cooling tower
which helps to decrease the temperature as the uid drops down through the cooling tower to be re-used in the condenser. Liquid
phase is driven by circulation pumps and air is drawn into the cooling tower by fans.
Generally, condensers are operated at the lowest temperature possible results in maximum turbine work and cycle efficiency and in
minimum heat rejection. The typical condensate temperature attained in practice is 45-50C, corresponding to a condenser pressure
of 9.6-12.5 kPa-abs (El-Wakil, 1984; Moghaddam, 2006).
3. METHODOLOGY
The general assumptions and constant parameters of representative single-ash GPP model are listed in Table 1.The
thermodynamic model uses the data of Kizildere Geothermal Power Plant (KGPP)-Turkey, which is a single-flash plant with
extremely high NCG fraction, to allow a comparison between the results of the modelling and the operational data of an actual
single-flash GPP.
For all processes of the power plant, CO2 is considered not to dissolve in the water.
Turbine efficiency is calculated according to Baumann Rule (DiPippo, 1982) and the calculation of isentropic quality considers
the existence of NCGs.
The temperature difference between cooling water entering the cooling tower and hot air leaving the cooling tower is 6C
(Siregar, 2004; Swandaru, 2006).
The temperature drop of the condenser exit to the cooling tower entrance is 3C (Swandaru, 2006).
The temperature of CO2 gas is assumed same as to the wet bulb temperature of cooling water (Swandaru, 2006).
NCG removal systems are considered as two-stage and each stage is assumed to use equal pressure ratios.
The pressure drop throughout the inter and after condensers is assumed as 1 kPa.
The general equations of mass and energy balance used in the static (steady-state, steady-flow) model are summarized in Table 2.
mm 1
in
m
x m
out
(4)
(5)
l
m s x 1 f m (6)
m NCG x f m (7)
m h m l hl m s hs m NCG hNCG (8)
m h m h
in out
(9)
Overall mass and energy balance for steady-state conditions with reference to Figure 1, can be expressed as below (Kwambai,
2005). The subscript numbers refer to state locations on Figure 1.
10 m
m air m
12 m
13a m
22 m
31 m
air (10)
A B
3
Yildirim and Gokcen
Figure 2 exhibits the input and output parameters of each sub-module. The sub-models work simultaneously using output
parameters of each sub-module as input parameters of the others. The main output of the module is net power output, total auxiliary
power of the plant and specific steam consumption of the plant.First step of the module is to determine the optimum separator and
condenser pressures which give the maximum net power output and minimum total auxiliary power. The main equations of the sub-
models except NCG Removal System Module are summarized in Table 3.
Compression is ideally an isentropic process. To determine the actual enthalpy at compressor exit is quite complex since the
geothermal steam is a mixture of water vapor and CO2. Therefore, the isentropic enthalpies of water vapor and CO2 are calculated
separately. Then, the isentropic enthalpy of the mixture is calculated using the mass flowrate of water vapor and CO 2 and their
isentropic enthalpies.
m s,17 h s,17,is m NCG,17 h NCG,17,is
h17,is (35)
m 17
4
Yildirim and Gokcen
The actual enthalpy at the compressor exit is calculated using Eq. 36.
h17,is h16
comp (36)
h17 h16
A steam jet ejector operates on the venture principle. The motive steam is expanded through the nozzle to the design suction
pressure. The pressure energy of the steam is converted to velocity energy and on leaving the nozzle at high supersonic velocities
the steam passes through the suction chamber and enters the converging diffuser or entrainment, as gas and associated water
vapor.Because of the capacity of a single ejector is fixed by its dimensions, a single unit has practical limits on the total
compression and throughout it can deliver. For greater compression, two or more ejectors can be arranged in series (Hall, 1996;
Swandaru, 2006; Birgenheier et al., 1993). In a multi-stage system, inter-condensers are typically used between the stages. By
condensing the vapour prior to the next stage, the vapour load is reduced. This allows smaller gas removal systems to be used, and
reduces steam consumption. An after-condenser can also be added, to condense vapour from the final stage. This will not affect
overall system performance, but may ease disposal of vapour and acts as a noise suppressor (Swandaru, 2006; Birgenheier et al.,
1993).
The suction and discharge pressure of each stage is determined by the following calculations (Geothermal Institute, 1996b). Each
stage uses equal pressure ratios based on system suction and discharge pressure of 90% condenser pressure and 105 kPa. SJEs are
feed by the motive steam, which leaves the separator. Between the stages the gas coolers are used. Daltons laws of partial pressure
and ideal gas equations are used to calculate necessary motive steam flowrate at point 33 and 34 (Hall, 1996).The air steam ratio
(AS) can be found by Air to steam ratio curve (Geothermal Institute, 1996b) that has been transformed into a small program in EES
called procedure ratio_1. Inputs required for this program are the expansion ratio and the compression ratio.
)
0.86
2.01 ( M NCG
E NCG 5.73 104 18.36
The entrainment ratio for NCG (38)
0.86
18.36 ( M NCG )
5
Yildirim and Gokcen
The flow diagram of HS which is a combination of SJE and LRVP is demonstrated in Figure 5.
1
1
m NCG Ru TNCG P19
W LRVP
1 (43)
1 LRVP M NCG P18
NCG,36 0.98 m
m NCG,13b (44)
s,36 m
m NCG,36 (45)
s,42 (m
m s,13b m
s,36 ) 0.01 (46)
6
Yildirim and Gokcen
condenser can also be added to condense vapor from the final stage. Adding an after condenser will not affect overall system
performance, but may ease disposal of vapor and acts as a noise suppressor. (Birgenheier et al., 1993; Swandaru, 2006).
4. RESULTS
For the given data of KGPP and the assumptions made, thermodynamic analysis is carried out and the impacts of separator pres-
sure, NCG fraction, wet bulb temperature on the net power output and specic steam consumption are discussed. Then, the results
are compared with the operational data of KGPP.
The main results of the mass and energy balance of the plant are presented in Table 5.
Table 5: Main results of the mass and energy balance of the plant with Kizildere operational data.
NCG Removal System CS SJES HS RS
Separator Pressure (kPa) 460 460 460 460
Condenser Pressure (kPa) 10 10 10 10
Compressor /LRVP 1262 1299
Auxiliary Steam Jet Ejector * 6666 3038 180
Power Water Circ.Pumps 346 372.4 360.3 192
Cooling Tower Fans 86.3 91.5 89.8 47.2
(kW)
Other 150 150 150 150
TOTAL 1844 7279 4936 569.2
Net Power Output (kW) 10235 5466 7447 5667
14000
12000
Net Power Output (kW)
10000
8000 Average: 9505 kW
6000
4000
2000
0
1984
1985
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
1997
1998
1999
2000
2001
2002
2003
2004
Year
Figure 7: Annual average net power output of Kizildere GPP (Source: DPT, 2001; Yildirim and Gokcen, 2004)
7
Yildirim and Gokcen
Figure 8: Net power output of the plant for various separator and condenser pressures at 13% NCG fraction.
Generally, GPPs operate at off-design conditions. To be able to compare thermodynamic performance of the plant with operational
and optimum separator pressures, net power output and auxiliary power of the plant are calculated at optimum separator pressure
for each NCG removal system and the results, which are summarized in Table 6, show that the net power outputs are increased as
0.2-11.7% by using optimum separator pressures. As an example, the net power output of the CS is calculated as 11436 kW at
optimum separator pressure of 220 kPa at 13% NCG fraction. An average operational separator pressure of KGPP is as high as 360
kPa. The decrease in net power output is approximately 1,2 MW (11,7%) because of the elevated operational separator pressure.
Table 6: Main results of mass and energy balances of the plant at optimum separator pressures.
NCG Removal System CS SJES HS RS
Optimum Separator Pressure (kPa) 220 500 340 580
Condenser Pressure (kPa) 10 10 10 10
Compressor /LRVP 1749 1518
Auxiliary Steam Jet Ejector * 6239 3645 370
Power Water Circulation 486 353 424 252
Pumps Tower Fans
Cooling 121 87 106 62
(kW)
Other 150 150 150 150
TOTAL 2506 6829 5843 834
Net Power Output (kW) 11436 5476 7712 6294
* Consumed motive flow rate is converted into power in kW.
Figure 9: Turbine power output, net power output and auxiliary power of the plant vs. NCG fraction.
In Figure 10separator pressure versus net power output of the plant for various NCG fractions (0-25% by weight of steam) is
demonstrated at 10 kPa condenser pressure.The Figure 10 indicates that each option exhibits the same behavior for zero NCG
fraction except RS. Because RS requires at least 330 kPa pressure drop between the separator and turbine inlet, while the other
NCG removal systems require 10 kPa.
8
Yildirim and Gokcen
Figure 10: Separator pressure vs net power output of the plant for various NCG fractions.
Figure 11 gives a better insight of the optimum separator pressures depending on NCG fraction. Increasing NCG fraction increases
optimum separator pressures for each NCG removal system.
700
Optimum Separator Pressure.
600
500
400
(kPa)
300
200
100
0
CS
0 5 10 15 20 25
HS
SJES
RS NCG Fraction (%)
12000 12000
CS HS
Net Power Output (kW)
SJES RS
9000 9000
6000 6000
3000 3000
CS HS
SJES RS
0 0
5 10 15 20 25 5 10 15 20 25
Wet Bulb Temperature (o C) Wet Bulb Temperature (o C)
Figure 12: Net power output and auxiliary power of the each system vs. wet bulb temperature.
9
Yildirim and Gokcen
Table 7: Wet bulb temperature vs net power output and auxiliary power of the plant.
As it can be observed from Figure 12, the net power output of the plant decreases with increasing wet bulb temperature. This is
because increasing the wet bulb temperature, increases the motive steam flowrate, since the auxiliary power increases. The results
of Table 6 is depicted that net power output is decreased as 0.18% for CS, 0.37% for SJES and HS and 0.04% for RS, while
auxiliary power is increased as 0.84% for CS, 0.33% for SJES, 0.55% for HS and 0.31% for RS by 1C increment in wet bulb
temperature.
16
CS
Specific Steam Consumption .
HS
12
SJES
(kg/s /MW)
RS
8
0
0 5 10 15 20 25
NCG Fraction (%)
4. CONCLUSIONS
The results of modeling and simulation of flashed- steam GPPs have been summarized. The model has been confirmed using data
of Kizildere GPP as input parameters. Simulation parameters are wet bulb temperature, separator pressure, condenser pressure and
NCG fraction.
Optimum condenser pressure for each NCG removal system is determined as 10 kPa.
Optimum separator pressures are determined as 220 kPa, 340 kPa, 500 kPa and 580 kPa for CS, HS, SJES and RS,
respectively.
Thermodynamic performance of single- flash plant can be improved by 0.2-11.7% running the separator and condenser
pressures on their optimum values. GPPs should be urged to operate around design conditions to generate optimum net
power.
Net power output is decreased by 0.18% for CS, 0.37% for SJES and HS and 0.04% for RS by 1C increment in wet bulb
temperature.
Specific steam consumption is the highest for RS and lowest for CS. As an example at 460 kPa separator pressure; for 2%
NCG fraction, RS consumes 47.4% more steam than CS, for 13% NCG fraction it is 97.4%.
1% increment in NCG fraction results a decrement on net power output as 0.4% for CS, 2.2% for HS, 2.7% for SJES and
2.5% for RS.
Based on the results of the thermodynamic modeling, CS is the best gas removal option in terms of the highest net power
output and lowest auxiliary power for Kizildere GPP operational conditions. On the other hand, RS is the worst option for
10
Yildirim and Gokcen
entire NCG fraction range. HS is responded late to the change in NCG fraction because the LRVP is more efficient since
its performance lies between CS and SJES.
REFERENCES
Awerbuch, L.; Van der Mast, V.; Soo-Hoo, R. Review of Upstream Geothermal Reboiler Concepts, Geothermal Resources Council
Transactions,8, (1984).
Barber Nichols Home Page. http://www.barber-Nichols.com/products/blowers_and_ compressors/turbocompressors/default.asp
(accessed May, 2010).
Bertani, R. Geothermal Power Generation in the World 20052010 Update Report, Proceedings of World Geothermal Congress,
Bali, Indonesia, April 25-29, 2010.
Bidini, G.; Desideri, F.; Di Maria, F. A Single Flash Integrated Gas Turbine - Geothermal Power Plant With Non Condensable
Gasses Combustion, Geothermics, 28, (1999), 1131-150.
Birgenheier, David B. et al. Designing Steam-Jet Vacuum Systems, Chemical Engineering, July 1993, www.graham-
mfg.com/downloads/23.pdf (accessed 2008).
Coury, G.; Guillen, H. V.; Cruz, D. H. Geothermal noncondensable gas removal from turbine inlet steam, Proceedings of the 31st
Intersociety 3 Energy Conversion Engineering Conference, 1996.
Coury, G.E. and Associates. Upstream H2S removal from geothermal steam, Technical Report EPRI, AP-2100, 3.1-3.23. 1981.
DiPippo, R. The Effect of Expansion-Ratio Limitations on Positive-Displacement, Total-Flow Geothermal Power Systems,
Geothermal Resources Council Transactions, 1982, 6, pp 343-46.
DPT, Sekizinci Be Yillik Kalkinma Plani, Technical Report of Madencilik zel htisas Komisyonu Enerji Hammaddeleri Alt
Komisyonu Jeotermal Enerji alima Grubu, DPT. 2609 K. 620, 2001 (in Turkish).
http://ekutup.dpt.gov.tr/madencil/enerjiha/oik620.pdf (accessed, 2010) (in Turkish).
Duthie, R.G.; Nawaz, M. Comparison of Direct Contact and Kettle Reboilers to Reduce Noncondensables in Geothermal Steam,
Geothermal Resources Council Transactions, 1989, 13, pp 575-580.
El-Wakil, Geothermal Plant Technology, McGraw-Hill Inc, NewYork, 1984, 859 pp.
F-Chart Software Home Page. http://www.fchart.com/ees/ees.shtml (accessed 2014).
Geothermal Institute, Gas extraction system, Course notes of Geothermal Institute, Auckland University, Diploma course in energy
technology (geothermal), 1996b, 75 pp.
Gunerhan, G. An Upstream Re-Boiler Design for Removal of NCGs from Geothermal Steam, Geothermal Institute Report, Report
No: 96.10, University of Auckland, New Zealand, 1996.
Hall, N. R. Gas extraction system, Lecture notes of Geothermal Utilisation Engineering Geothermal Institute, The University of
Auckland, New Zealand in M.G. Dunstall (eds.), 1996.
Hankin, J. W.; Cochrane, G. F.; Van der Mast, V. C. Geothermal Power Plant Design for Steam with High Noncondensable Gas,
Geothermal Resources Council, Transactions, 1984, 8.
Michaelides, E. E. The influence of Noncondensable Gases on the Net Work Produced by the Geothermal Steam Power Plants,
Geothermics, 1982, 11-3, pp 163-174.
Moghaddam, A.R. A conceptual Design of A Geothermal Combined Cycle and Comparison with a Single-Flash Power Plant for
Well NWS-4, Sabalan, Iran, Report of the United Nations University Geothermal Training Programme, Reykjavik, Iceland,
Report No:18, Pp:391-428, 2006.
Montero, G. Evaluation of the network of a turbine operated by a mixture of steam and non-condensable gases, Proceedings of 12th
New Zealand Geothermal Workshop, 1990, 11, pp 163174.
Siregar, P. H. H. Optimization of Electrical Power Production Process for the Sibayak Geothermal Field, Indonesia, Report of the
United Nations University Geothermal Training Programme, Reykjavik, Iceland, Report No:16, Pp. 349-76, 2004.
Swandaru, R. B. Thermodynamic Analysis of Preliminary Design of Power Plant Unit I Patuha, West Java, Indonesia, Report of the
United Nations University Geothermal Training Programme, Reykjavik, Iceland, Report No:7, Pp. 83-119, 2006.
Tajima, S.; Nomura, M. Optimization of Non-Condensable Gas Removal System in Geothermal Power Plant, Geothermal Resource
Council Transactions, 1982, 6, pp. 397-400.
Vorum M.; Fritzler E.A. Comparative Analysis of Alternative Means for Removing Non-Condensable Gases From Flashed-Steam
Geothermal Power Plants, Report of NREL SR-550-28329, Colorado, The USA, 2000.
Yildirim, E. D.; Gokcen, G. Exergy analysis and performance evaluation of Kizildere geothermal power plant, Turkey, Int. Journal
of Exergy, 2004, 1-3, pp 316-333.
NOMENCLATURE
AS : Air-steam ratio (-)
11
Yildirim and Gokcen
Greek letters
: Efficiency (%)
: the specific volume of the water vapor (m3/kg)
P : Pressure drop (Pa)
: Cp / Cv (-)
T : Temperature difference (C)
: Humidity ratio (-)
Subscripts
12